building production infrastructurebuilding custom softwarebuilding custom software, production infrastructure, security research, health-tech platforms, software that just works
A one-person software company. Software built end to end and run in production, and security research that finds real flaws in live systems.
All of it built and run solo. What Zyric builds and offers first, then the rest of the shipped work.
In Development
Foundation
Flagship
A control plane for servers and applications: one audited API for every change, honest status, sealed and verified backups, and active defence at the edge.
Runs production systems for external companies and most of my projects, including this site
Runs most of my projects in production, including this site. Not sold as a public product: client projects can run on it, app, database, storage and backups on one server I set up.
Foundation is Zyric's flagship product: a control plane for servers and applications, built on three rules: run it honestly, act through one door, prove everything. It runs the software that already works (Docker, Postgres, Redis, Caddy, Git and SaaS) rather than replacing it. Every project, repo, service and database is a resource with a desired and an observed state, and controllers converge what is running toward what was declared: you change desired state instead of clicking through consoles. Every change, by a person, a script or an agent, goes through one checked and audited API, and no status turns green unless it is: a failed read is never an empty answer. Backups are sealed on the box, verified after every run and restored in a trial run; scanners probing the edge are banned and caught with bait keys; and Repository Intelligence reads every repository with fixed rules first, with AI analysts only on request. It is still in active development and already runs production systems for external companies and most of my projects, including this site.
Control Plane · AI · Infrastructure · Security · MCP
In Development
Tessika
Privacy by Construction
Any document, photo or scan in, one clean structured record out, with the personal data on it found and masked first.
Health, legal, finance, insurance and identity · self-hosted
Working end to end and audited against documents with known planted personal data. Offered on request, adapted to your documents.
Tessika reads documents the way a person would, straight from the image rather than through OCR: PDFs, scans, phone photos, handwriting in many cases, Word and Excel files, and plain text. It finds the personal data on the page (names, dates of birth, addresses, ID and account numbers, and on the strict setting faces, signatures and the machine-readable zone of a passport), masks it, then re-reads the masked page and covers anything still legible. What comes out is one canonical record per fact, with a confidence and a trail back to its source, in the shape the system above it asks for. Health, legal, finance, insurance and identity documents run through the same engine; a new domain is a plugin, not a rebuild. It runs in the customer's own environment on an open-source model that can be swapped, so no document has to leave for a third-party AI service, and it ships with its own audit: documents with known planted personal data, and a leak check that reads the masked output back.
Privacy · Documents · Redaction · Self-hosted · Data Infrastructure
A security-review pipeline that reviews critical code the way a specialist team would, invariant by invariant, across smart contracts, bridges, light clients, ZK circuits, and the web2 systems that guard identity, authentication, and access-control. To date: 100+ targets reviewed across 30+ ecosystems against 200+ vulnerability classes, with 20+ confirmed findings, every one human-triaged and severity-calibrated by a human, not the model alone. That calibration is the point: in a live cross-chain bridge that has moved $100M+ across its routes, the pipeline flagged real consensus-verification gaps in the non-flagship light clients as critical; a person traced the entry point and corrected the rating to conditional, medium, but the gap between the bridge's marketed trustless verification and what those clients actually check still stood (responsibly and privately disclosed). Detection-only: it finds and reports, never exploits. One engine spanning EVM, Solana, Move, Cosmos, Substrate, Bitcoin, ZK circuits, and web2 access-control.
A free, MIT-licensed scanner for the person a suspicious take-home task was sent to. Fake-interview campaigns hide a payload in a postinstall hook, an obfuscated config, or a dependency pinned deep in a lockfile, and it runs the moment you type npm install: browser passwords, SSH keys, cloud tokens and wallets, gone before you have read a line. RepoCanary never clones, installs, executes or evaluates the target. It reads a bounded set of files as plain text through the GitHub API and matches signatures, so pointing it at malware is exactly as dangerous as reading it in a browser. Detection is measured against a corpus of the techniques these campaigns actually use, and the evasions it cannot catch are written into the threat model rather than hidden. Green means nothing known matched, never safe, and the tool never says anything stronger. One command, zero runtime dependencies, and one shared engine behind the CLI, the website and a GitHub Action. The code is open-source on GitHub.
Agents fail in ways a green build never catches: a final answer that promises work nothing will ever do, an invented "I can't access that" with no tool tried, "saved a memory" when no memory was written, a count quoted from memory instead of from the listing that was just made, a list cut short without anyone being told. None of those are model bugs; they are bugs in the loop around the model. looptape records a real session as a tape (every completion, tool call and result, exactly as the model saw it), replays it into your own harness in place of the model, with no model and no network, and runs its checks over what happened. Every check comes from a real incident, and its origin is kept on the check. A recording proxy for the Anthropic and OpenAI APIs, a deterministic player, zero dependencies, and nothing in it needs an API key. The code is open-source on GitHub.
The complete digital surface of a Vietnamese street-food restaurant in Schwäbisch Gmünd, built end to end and shipped solo. The public site carries the menu (allergens, vegan and vegetarian markers, spice levels, sold-out for today), opening hours, online ordering with pickup slots, reservations, and events. Behind it sits an eleven-tab dashboard where the owners edit all of it themselves, with no deploy and no developer in the loop: dishes, prices, opening hours, holidays, the editorial copy, and the FAQ. Orders and reservations arrive by mail and Telegram and stay in the dashboard, with each path standing on its own: if the database is down the mails still go out, and if the mail provider is down the order is still in the dashboard. Next.js 16, React 19, Tailwind v4, Drizzle and Postgres.
A full-stack governance client for Enjin OpenGov, all backed by live chain RPC. Browse referenda, cast conviction votes, file on-chain proposals end to end, comment, and manage a profile, straight from the browser. Multi-wallet support (Enjin Wallet, WalletConnect, and the four major Polkadot extensions), a live and an archive node for Enjin Relaychain and its Canary testnet, zero required indexers, SIWE-style sign-in by signature, and the EGOV1 on-chain metadata standard that pins every proposal's text to the chain by hash, so anyone can check it for edits and any third party can rebuild the proposal corpus. Version 2.0 adds proposals written for voters (Markdown with images and address chips, verified against their EGOV1 record), admin calls such as runtime upgrades, the Fellowship with rank-weighted voting, and moderation with roles, a review queue and a public log. The client is open source on GitHub.
Substreak is a Substrate reward and asset-distribution system. It began with a Telegram interface; rather than run custodial wallet infrastructure on users' behalf, that user-facing layer was wound down by choice. The backend never stopped: automated reward distributions run on schedule, and the integration APIs stay live for the systems that depend on them. A revival is in the works, rebuilt from the ground up and built solo, for launch in 2027.
nimimo is a non-custodial crypto identity layer. One link (nimimo.com/@chris) receives Bitcoin, Ethereum, Solana, and USDC, abstracting wallet complexity away from the user. The architecture deliberately separates access, identity, ownership, and recovery, with keys remaining fully on-device. Built for security-critical flows, cross-chain usability, and reducing friction in real-world crypto transactions, with no app, no exchange, and no KYC. Ships with a lightweight, no-API-key developer SDK.
Crypto · Identity · Non-custodial · Multi-chain · Open Core
Private
OneHealth
Co-Founder & CTO, 2026
The operating layer and navigator for personal health: one lifelong record, and a clinician-approved route through it.
Co-Founder & CTO, first version built solo
Proprietary. I designed and built the first version; further development is with OneHealth.
OneHealth is the system of record for a person's health, with a navigator built on top of it. Every input (lab document, wearable export, clinician's note, pharmacy slip, image) is normalised into one lifelong, source-agnostic record, where each value carries its unit, reference range, timestamp, source and quality flag, and nothing is ever overwritten. The product sits on that record: you set a destination (lose fat, lower cardiovascular risk, sleep deeper, age slower), the system computes a ranked, reasoned route, a licensed clinician approves it, and it recalculates every time your data changes. The system proposes; a human stays accountable. Built on an event-sourced canonical layer, an AI agent council, and a clinician review loop.
Health Tech · AI · Privacy · Longitudinal · Clinical
A solo worldbuilding project exploring procedural generation, persistence, and discovery. Every world is generated deterministically from a single seed, producing its geography, climate, weather, life, and sky. A seed is a place rather than a session, meaning worlds can be revisited, shared, and explored together while remaining identical for everyone who visits. Build a homestead, light a watchfire, plant your standard, and leave a mark on the world. Time passes, weather changes, moons drift overhead, and every world develops its own character from the same underlying rules. Built with React, TypeScript, React Three Fiber, and a fully procedural audio system. No accounts, no downloads, and no backend required.
Foundation runs the servers; Zyric Console runs the company. It is the private app behind zyric.de, installed on my phone's home screen and signed in with a passkey. It edits this website (projects, texts and skills, with every save kept as a revision), collects contact requests from the site and from the /k contact card, and sends mail from my own address with its delivery status. The Ablage keeps every file in private storage, with time-limited share links and AI actions that file, describe or transcribe a document on request. The business side writes offers, order confirmations and invoices ready to send; a customer accepts an offer online with a signature, and the signed copy is filed automatically. The tax view computes income and expenses by the day the money actually moved and runs fixed checks over them, and a model only phrases the advice on top of those numbers. Everything that happens reaches me as a push notification, with Telegram as the fallback, and Foundation's server reports arrive the same way. Built solo, used every day, and not for sale.
A snowed-in Lapland village in the browser: a giant fir, string lights, log cabins and the northern lights. Every visitor comes in as an elf and meets the others.
2.4 × 2.4 km of world, shared live with every visitor
In development: you can already walk around at christmas.zyric.de while the rest of the village is being built.
A solo multiplayer world for the Christmas season. You arrive at dusk in a village in Lapland, and it stays the blue hour: a 40-metre fir hung with string lights on the square, snowed-in log cabins with warm windows you can step into, a Christmas market with a Ferris wheel, a railway through the forest, and the northern lights overhead every evening. Every visitor comes in as an elf, and you meet the others in the same village: chat, whisper, throw snowballs, bake gingerbread and give it away. The terrain measures 2.4 by 2.4 kilometres, and places are only built as you come near them and taken down again behind you. It grew out of A Million Years and runs on the same engine: React, TypeScript, React Three Fiber, three.js and Vite, with a Node.js server whose WebSocket relay connects the players. The village is still being built.
Weekend builds and things I tinker with, all live. Swipe through.
01Live
Enjin City
2026
A walkable 3D WebGL gallery, built in a weekend. It preserves my old NFT Enterprise collection (many of its NFTs now lost) and renders other collections (not mine) as explorable buildings across the city. A browsable museum of NFT art.
Reads your heart rate and an HRV estimate from your phone camera (PPG), entirely on-device. The solo, duel, and tournament modes are a playful layer over research I did for OneHealth into low-cost health-data capture. iPhone-optimised.
A just-for-fun music side-quest. The songs are AI-generated, but I direct the composition and do the sound engineering, and every release is mastered through a self-built pipeline (around 50 tracks, including a full Christmas album). Making good songs is easy; the music industry isn't, so this stays a playground.
An early social NFT platform in the Enjin ecosystem: 30+ artists, tens of thousands of holders, and QR-based distribution and onboarding. Its collections are still explorable inside Enjin City, a walkable 3D gallery across 40+ floors.
A multisensory state-design studio: sound, rhythm, body vibration (Woojer haptics), breath, and light turned into designable, measurable sessions, built on a strict honesty model that keeps the measurable, the hypothesised, and the traditional clearly apart and never dresses up the unproven as fact.
Kinetara
Ongoing · Private
A headless crypto market signal engine: no interface, just a cron and APIs. It scores live perpetual-futures data each hour and broadcasts a long, short, or stay-silent call to Telegram.
Falk
Ongoing · Experiment
An AI authorship engine for entire literary universes: world-building, characters, lore, and chapters, writing internally consistent multi-book trilogies. A fun benchmark for how far AI world-building can go.
NFT Tipper Bot
2023-2025 · Discontinued
A Telegram-based NFT distribution system on Enjin that processed 100k+ on-chain transactions in production. Its scaling limits are what pushed me toward non-custodial infrastructure.
About
Oneperson,thewholestack.
Founder, engineer, and the entire team in between: from the first blank repo to the thing that runs in production.
I run Zyric, a one-person software company: I build software end to end, run it in production, and do security research. My flagship is Foundation, the control plane most of my projects run on, including this site. It isn't sold as a public product, but client projects can run on it too: the app, its database, storage and backups on one server I set up, instead of a separate bill for every managed service. My research line is Panoptes, an automated security-review pipeline that's reviewed 100+ targets across 30+ ecosystems and flagged real consensus-verification gaps in a live cross-chain bridge that's moved over $100M. Alongside that I built RepoCanary, a free open-source scanner that checks a take-home task for fake-interview malware without ever running it, and as co-founder and CTO of OneHealth I built its first version on my own, the event-sourced canonical record behind a health navigator. Before and alongside all of this I've spent around a decade in crypto, shipping the rails rather than the hype: Enjin Governance is open source, because I like giving something back to the Enjin ecosystem, and nimimo's cryptographic core is open under AGPL-3.0. Enjin Governance 2.0 is out, and a full revival of Substreak is planned for 2027. Everything here is built and run solo, end to end. I also open-sourced looptape, a tool that records an AI agent once and proves on replay whether it told the truth. If any of that maps to what you're building, whether that's software built end to end, infrastructure and operations, a security review, a consulting engagement, an investment, or just a conversation, I'd like to hear from you.
Now
Running Panoptes and RepoCanary, and taking on engineering, infrastructure and security work.
Open to
Software built end to end, infrastructure and operations, security reviews, or consulting.